This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Role of Contrast Enhanced Spectral Mammography to Predict Upgrade Rates of Biopsy Proven Atypical Ductal Hyperplasia (NCT03505372), a na na study. The primary outcome is CESM's Ability to Predict Upgrade Rates of Biopsy Proven ADH at Surgical Excision.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03505372 |
| Phase | NA |
| Condition | Atypical Ductal Hyperplasia |
| Allocation | Not applicable (single group) |
| Design model | Single Group |
| Masking | None (open label) |
| Enrollment | 5 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Dana-Farber Cancer Institute |
| Results posted | Yes |
2. Clinical Question
Population
People with biopsy-proven atypical ductal hyperplasia (ADH) who were being evaluated for surgical excision. The registry record lists atypical ductal hyperplasia as the condition and describes the primary question in relation to upgrade at surgical excision. It does not report further eligibility details or participant characteristics.
Intervention
Contrast enhanced mammography (CESM). The study used CESM images, with two radiologists described as prospectively reviewing them. The registry record does not report enough detail to characterize the reviewers’ scoring system: the arm description ends before that information is complete.
Comparator
No separate comparator group is listed. This is a single-group study, not a randomized comparison. Consequently, it does not provide a like-for-like control group against which to compare CESM or its use in clinical management.
Primary question
Can enhancement on CESM predict which cases of biopsy-proven ADH will be upgraded at surgical excision? The registry record specifies a primary-outcome timeframe of 3 months. It also states that images were performed, but outcome data were not collected, so the prediction question cannot be answered from the posted results.
3. Trial Design
NCT03505372 is a single-group study of contrast enhanced mammography in people with atypical ductal hyperplasia. The listed intervention is the CESM device, and the registry record describes the purpose as examining whether CESM can predict upgrade to cancer. Allocation is recorded as NA, and the record identifies no masking. This is not a randomized trial or a comparison of CESM with a control intervention.
The study enrolled 5 participants and is listed as terminated. Its recorded start date is 2018-05-10 and its primary completion date is 2019-06-28. The phase is listed as NA. Dana-Farber Cancer Institute is identified as the sponsor. The registry record does not explain why the study was terminated or report a planned enrollment target.
The intervention description says that CESM images were to be obtained according to clinical protocol, within approximately 2-12 minutes of contrast injection, with four images per breast using low and high energy. It describes prospective review by two radiologists. The remaining review description is incomplete in the registry record, so the precise image interpretation procedure cannot be characterized from that text.
The central design limitation is that the intended prediction outcome was not collected. Although imaging was performed, the record explicitly says that outcome data were not collected and that no analysis can be provided. The study therefore supplies no estimated association between CESM enhancement and surgical upgrade, and it cannot establish how well the imaging finding discriminates between cases that do and do not upgrade.
4. Participant Flow and Accounting
The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.
Overall Study
| Stage | Contrast Enhanced Mammography |
|---|---|
| STARTED | 5 |
| COMPLETED | 5 |
| NOT COMPLETED | 0 |
Analysis population for the primary outcome, as defined in the registry: Images were performed but outcome data was not collected and therefore we cannot provide any analysis.
5. Baseline Characteristics
The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.
Population: Patients who had ADH and were getting surgical excision were enrolled. We terminated the study as recruitment was too low to continue.
| Characteristic | Unit | Contrast Enhanced Mammography |
|---|---|---|
| Participants analysed | 5 | |
| Age, Categorical: <=18 years | Participants | 0 |
| Age, Categorical: Between 18 and 65 years | Participants | 5 |
| Age, Categorical: >=65 years | Participants | 0 |
| Age, Continuous | years | 50.2 (47 to 56) |
| Sex: Female, Male: Female | Participants | 5 |
| Sex: Female, Male: Male | Participants | 0 |
| Race and Ethnicity Not Collected | Participants | — |
| Region of Enrollment: United States | Participants | 5 |
6. Primary Endpoint
The registry lists the primary outcome as CESM's Ability to Predict Upgrade Rates of Biopsy Proven ADH at Surgical Excision, measured over 3 months.
| Element | Registry entry |
|---|---|
| Result type | N/A |
| Measure of spread | not applicable |
| Unit | not stated |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint concerns CESM’s ability to predict upgrade of biopsy-proven ADH at surgical excision. In this context, upgrade refers to a change in the diagnosis identified at excision. The registry describes the intended assessment as evaluating whether enhancement on CESM can predict which ADH cases upgrade. The stated timeframe is 3 months.
As a general statistical concept, a prediction endpoint is useful when a study aims to determine whether information available before a later clinical finding can distinguish people with different outcomes. A meaningful assessment ordinarily requires both the imaging result and the outcome for the relevant participants. Depending on the study’s prespecified approach, performance might be described with measures of discrimination, calibration, or classification accuracy. The registry record does not report which specific performance measures were planned.
Here, the endpoint was not analyzed because outcome data were not collected. No number of upgrades, prediction estimate, uncertainty interval, or performance measure is reported in the registry record. The endpoint’s clinical purpose can therefore be described, but its observed performance cannot be estimated from these results.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The posted results do not answer whether CESM enhancement predicts upgrade of biopsy-proven ADH at surgical excision. The registry record states that images were performed but outcome data were not collected; it consequently reports no analysis of the primary outcome. There is no reported estimate of prediction performance, no uncertainty interval, and no result that supports interpreting CESM findings as more or less informative about upgrade in this study.
The study enrolled 5 participants, but enrollment alone is not an outcome result. Without outcome information linked to the imaging findings, the number enrolled cannot establish how many participants experienced upgrade or whether enhancement distinguished cases with different outcomes. No upgrade count or participant-level pairing of imaging and surgical findings is reported in the registry record.
The single-group design also has no separate control group. That is consistent with a prediction question, which need not always be framed as a treatment-versus-control comparison, but it does mean that the study cannot estimate a comparative effect of CESM against another imaging strategy or management approach. The registry record does list a secondary outcome concerning types of changes in surgical management based on CESM, but it reports no analysis for that outcome either.
As a general statistical concept, prediction claims depend on observing both predictors and outcomes. When outcome information is absent, measures of discrimination and calibration cannot be estimated from the study record. Even if images were read prospectively, that fact by itself does not show whether the readings corresponded to surgical findings. The record gives no sensitivity, specificity, predictive values, or other prediction metrics.
The absence of an analysis should not be interpreted as evidence that CESM does or does not predict upgrade. It means the registry record contains no outcome-based estimate with which to assess that question. Nor can the results establish an absence of upgrade or demonstrate that the imaging approach changed management. The study status is listed as terminated, but the reason for termination is not reported in the registry record.
The posted safety figures describe events among the enrolled group, but they do not fill the missing efficacy or prediction data. Safety counts and prediction performance address different questions. The interpretation of the prediction objective must therefore remain limited to the design intent and the documented absence of outcome data.
9. Statistical Methods Explained
What kind of analysis would address the stated question?
The primary objective is predictive: whether CESM enhancement is informative about upgrade at surgical excision. As a general statistical concept, this requires an imaging result and a corresponding outcome for each participant included in the analysis. The statistical method then depends on how enhancement and upgrade are defined and on the intended use of the prediction. The registry record does not report the planned definitions or a statistical analysis plan.
For example, if both the imaging assessment and the surgical outcome are categorized, an analysis could compare the frequency of upgrade across imaging categories or evaluate how well a prespecified imaging classification identifies upgrades. If the aim is to estimate probabilities rather than simply classify cases, calibration would also matter: predicted probabilities should be compared with observed outcome frequencies. These are general methodological possibilities, not analyses reported for this study.
Why is there no prediction estimate?
The registry record says that images were performed but outcome data were not collected, and that no analysis can be provided. The missing outcome information prevents the required comparison between CESM readings and surgical findings. It is therefore not possible to calculate prediction performance or assess the precision of such an estimate from the posted results. No result should be inferred from the fact that imaging took place.
In general, a confidence interval expresses the uncertainty around an estimated quantity under the assumptions of the analysis. It cannot be constructed for the study’s prediction performance when no such estimate is reported. The registry record does not report confidence intervals, hypothesis tests, or other inferential results for the primary endpoint.
What does the study design contribute?
The study is described as single-group, with allocation listed as NA and no masking. It is not a randomized comparison, and no separate control arm is reported. For the prediction objective, the key evidence would be the relationship between the imaging assessment and the outcome, rather than a randomized contrast between groups. Even so, a prespecified reading rule, complete outcome ascertainment, and transparent reporting of uncertainty are important for judging how reproducible and useful a prediction result might be. Those details are not reported in the registry record.
The arm description says that two radiologists would prospectively review CESM images. The available description does not fully state how their assessments were classified or combined. As a general matter, agreement between readers can affect the reproducibility of an imaging-based predictor. No reader-agreement result is reported in the registry record, so reproducibility cannot be evaluated from these data.
How should planning illustrations be understood?
The stated sample-size assumptions describe a planning illustration based on an observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The assumptions also say that an effect to detect is expressed as a stated fraction of the observed effect, and distinguish between total events for time-to-event planning and participants per arm for other outcomes.
These are planning assumptions, not a description of this study’s design. The trial is single-group, and the registry record reports no primary-outcome analysis from which an observed effect could be obtained. No sample-size justification for the actual study is reported in the registry record. Any planning illustration should not be read as observed power or as evidence that the study had a particular probability of detecting a true effect.
As a general concept, over-dispersion occurs when observed variability exceeds that assumed by a statistical model. It can make standard errors too small if the model does not allow for the additional variability. The registry record does not report a model or an over-dispersion assessment, and the missing primary outcome means that no such assessment can be made from the posted results.
What can be inferred from the reported counts?
The enrollment and safety counts are descriptive. They do not supply the missing primary outcome. The registry record reports no analysis entries, and there is no reported estimate to interpret as statistically significant or not statistically significant. A lack of a reported test result is not a null result: it is an absence of an analyzed endpoint.
In general, estimates from small groups can be imprecise, especially when outcomes are uncommon. This study enrolled 5 participants, but because no primary outcome data were collected, it is not possible to quantify the precision of a prediction estimate or determine how many outcome events would have contributed to it. A confidence interval or a power calculation cannot substitute for the absent outcome observations.
10. Multiplicity, Missing Data and Analysis Populations
The registry record lists 2 outcome measures: the primary CESM prediction outcome and a secondary outcome concerning types of changes in surgical management based on CESM. It reports no analyses. Consequently, there are no posted outcome estimates or tests for which to assess multiplicity adjustments, and no adjustment strategy is reported in the registry record.
As a general statistical concept, multiplicity arises when several outcomes, comparisons, or analyses are examined. If statistical conclusions are based on multiple tests, the chance of at least one apparently notable result can increase unless the analysis plan addresses that issue. The record does not report a multiplicity plan, so it is not possible to determine whether any adjustment was intended. Since no outcome analyses are posted, multiplicity does not explain or resolve the absence of results.
Missing data are directly relevant to the primary endpoint: the registry record states that images were performed but outcome data were not collected. It does not report how many outcome values were missing, why they were missing, or whether any were available for particular participants. No missing-data method or sensitivity analysis is reported. Without surgical outcome data linked to the images, the primary prediction analysis cannot be conducted as described.
The registry record does not report analysis populations, such as a defined group of participants included in an endpoint analysis. It also does not describe exclusions or the handling of participants with incomplete records. These details cannot be inferred from the enrollment count alone.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Contrast Enhanced Mammography | 0 | 5 | 0.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Contrast Enhanced Mammography | 0 of 5 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Contrast Enhanced Mammography | 0 of 5 |
The registry record lists 5 participants at risk in the Contrast Enhanced Mammography group. It reports 0 participants affected by serious adverse events, 0 affected by other adverse events, and 0 deaths. No top serious adverse event terms are listed.
These are reported counts, not estimates of comparative risk. The study has no separate control group, and the registry record does not provide a comparison against another intervention or a background rate. The counts therefore describe only the group and event information recorded for this study; they cannot establish a comparative safety effect.
As a general statistical concept, a count of zero observed events does not establish that an event cannot occur. The interpretation depends on how events were defined, the completeness and duration of observation, and the amount of information collected. The registry record does not report the safety follow-up duration, detailed event definitions, or additional safety-analysis methods. The listed zero counts should not be read as proof that no adverse events were possible or that every potential safety outcome was assessed.
12. What the Result Does, and Does Not, Mean
The study addresses a clinically relevant prediction question: whether enhancement on contrast enhanced mammography can help identify biopsy-proven ADH cases that upgrade at surgical excision. It documents that CESM imaging was performed and describes prospective image review by two radiologists. Those details establish the intended approach, not its predictive performance.
The primary outcome data were not collected, and the registry record reports no analysis. As a result, the posted results do not show whether enhancement was associated with upgrade, how accurately it classified cases, or how uncertain any estimate might be. They also do not demonstrate that CESM changed surgical management, despite the listed secondary outcome about changes in management.
The result should not be framed as either a positive or a negative finding on prediction. No reported outcome estimate supports either conclusion. The terminated status likewise does not explain why the study ended; the reason is not reported in the registry record. The absence of results is not evidence that the proposed approach succeeds, fails, or has no effect.
As a general statistical distinction, a study can have a clearly stated research question and still lack the outcome observations needed to answer it. This record illustrates that distinction: the objective and intervention are described, but the essential endpoint information is absent. The enrollment and safety counts cannot replace outcome-linked imaging data.
The findings therefore support only a narrow reading: the study enrolled 5 participants, CESM images were performed, and the intended prediction outcome was not analyzed because outcome data were not collected. Any broader conclusion about prediction accuracy, comparative performance, or effects on care is not supported by the registry record.
13. Why This Trial Matters Statistically
This study matters statistically because it makes the difference between completing a measurement procedure and obtaining analyzable outcome evidence especially clear. CESM images were performed, but the registry record says that outcome data were not collected. A prediction study needs the imaging information and the outcome it is intended to predict; without both, the central relationship cannot be estimated.
The record also illustrates why design labels and planning assumptions must be interpreted in context. This is a single-group study, not a randomized comparison, while the stated planning assumptions describe equal allocation between two arms. Those assumptions are not a description of the actual trial design. The registry record does not report a sample-size justification for the study, and it reports no primary analysis from which an observed effect could be assessed.
As a general lesson in biostatistics, clear endpoint definitions, complete outcome ascertainment, and a stated analysis plan are important for evaluating predictive performance. Details such as how image readings are classified, how multiple readers’ assessments are handled, and which performance measures are used can determine what a result means. The available registry description does not fully report these methods, and no prediction analysis is available.
The record is therefore informative about the limits of what can be concluded from a trial listing. It identifies the question and the imaging intervention, while also documenting that the necessary outcome analysis is absent. That distinction helps prevent enrollment, imaging activity, or a zero-event safety count from being mistaken for evidence about the primary prediction objective.
14. Limitations
- The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
- Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.
The main limitation is that the primary outcome was not collected, despite imaging being performed. The registry record consequently reports no analysis of whether CESM enhancement predicts upgrade at surgical excision. There are no reported upgrade counts, prediction estimates, or measures of uncertainty.
The study enrolled 5 participants and is listed as terminated. The record does not explain the reason for termination or report a target enrollment or sample-size justification. With no primary outcome analysis, the contribution of the enrolled group to the prediction question cannot be evaluated.
The study has no separate comparator group and is not a randomized comparison. This prevents comparative conclusions about CESM versus another imaging approach or management strategy. The record lists a secondary outcome on changes in surgical management, but no results for it are reported.
The description of image review is incomplete. It states that two radiologists would prospectively review CESM images but does not fully report the scoring procedure, how reader assessments were combined, or whether agreement was assessed. The statistical analysis plan, analysis populations, handling of missing outcomes, and multiplicity approach are not reported in the registry record.
Safety reporting is limited to the listed event counts. The registry record does not report detailed safety definitions or the duration of safety observation. These limitations restrict interpretation of the event counts as well as the primary prediction question.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Role of Contrast Enhanced Spectral Mammography to Predict Upgrade Rates of Biopsy Proven Atypical Ductal Hyperplasia — NCT03505372.
- Registry record last updated 2020-02-11; status last verified 2020-01.
- Results first posted 2020-02-11.
- Data retrieved from ClinicalTrials.gov on 2026-11-09.
- Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
NCT03505372 is a terminated, single-group study sponsored by Dana-Farber Cancer Institute. It enrolled 5 participants and evaluated contrast enhanced mammography in relation to upgrade of biopsy-proven atypical ductal hyperplasia at surgical excision. The recorded start date is 2018-05-10, and the primary completion date is 2019-06-28.
The primary outcome timeframe is 3 months. The registry record states that CESM images were performed but outcome data were not collected, so no primary analysis is reported. A secondary outcome concerns changes in surgical management based on CESM, with no analysis reported for that outcome. The record lists 0 participants affected by serious adverse events, 0 affected by other adverse events, and 0 deaths among 5 at risk.